The bullet holes detection using image difference technology has the problems of over-detection, leak-detection, being sensitive to noise interference and harsh request for imaging condition. A bullet hole object detection method based on wavelet transform using fusion rules was proposed to overcome the limitations. In the wavelet domain, different fusion rules were applied to the low-frequency and the high-frequency sub-images to revise the wavelet transform coefficients, for increasing the object coefficients while decreasing or resetting the non-object coefficients. Then, it used the maximum points to the low-frequency coefficient and reduced the noise of high-frequency coefficients. The final bullet holes images were obtained through invert wavelet transform. Experimental results show that : the method can overcome the shortcomings of the image difference method, obtain images of bullet holes with clear edge, and improve the veracity and precision of the bullet holes detection.